Evaluation of Security Metric for Artificial Internet of Things Smart Locking System

• Main Author: Siti Nur Hanani, Mohd Rizal
• Format: Final Year Project / Dissertation / Thesis
• Published: 2023
• Subjects: T Technology (General); TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://eprints.utar.edu.my/5958/1/Siti_Nur_Hanani_Binti_Mohd_Rizal_21AGM06711.pdf; http://eprints.utar.edu.my/5958/

The increasing significance of delivering a smart lock system continues to grow. Modern technologies are being explored and utilise to accomplish this objective. Nonetheless, recent years have witnessed a noticeable surge in security breaches targeting the IoT systems, raising concerns for consumers. Drawing attention to this issue, the need to prioritise security should be considered during the development of the system, rather than treating it as an afterthought. The primary goal of this dissertation is to evaluate the security metric for IoT based locking system and design an enhance smart locking system that leverages the capabilities of Artificial Intelligence for facial recognition from the ground up. This system employs the Raspberry Pi controller to grant access to a private premises without relying on a physical keys like access cards. The designed smart lock system incorporates smartphones, enabling the sending of emails to the owner upon detecting a potential intruder. The SMTP protocol library is evaluated to enable the email transactions features from the microcontroller. In terms of hardware implementation, a microcontroller is chose, and a testing environment is established to explore and enhances security measures. This thesis outlines the architectural model of the smart locking system, the design of the face recognition algorithm, and the development of face recognition reader using the Raspberry Pi 3 model B microcontroller. Throughout this process, all developed sub-hardware components are design to interact seamlessly between one another, ensuring a functional and secure flow within the AIoT framework.